Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    money, cost, or dates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     agrade
    0.50
     collega
    0.48
    ष्मान
    0.48
    ดวก
    0.48
    赊
    0.47
    ة
    0.46
     AliExpress
    0.46
    प्लाई
    0.46
     мероприя
    0.46
     знать
    0.46
    POSITIVE LOGITS
    （
    0.57
     and
    0.55
    and
    0.53
    create
    0.49
     Function
    0.49
     which
    0.48
    while
    0.48
     producing
    0.47
    string
    0.46
    which
    0.46
    Act Density 0.018%

    No Known Activations